Development of a Cherenkov timing detector for measuring high-intensity secondary beams

K. Shirotori and T. Akaishi

for the E50 collaboration

Research Center for Nuclear Physics (RCNP)

Osaka University

3rd Symposium on Clustering as a window on the hierarchical structure of quantum systems

18th May 2020

Contents

• Introduction• Effective degree of freedom of hadrons: X* & W* spectroscopy

• High-momentum K- beam @ J-PARC High-p beam line

• High-rate Cherenkov timing detector: R&D status• High-intensity beam measurement by fine-segmented detector

• Time resolution evaluation of proto-type detectors

• To do

• Summary

2

Introduction

Investigation of effective degree of freedoms

• To understand role of effective degree of freedoms for hadrons• Diquark correlation, molecule states

• Systematic studies: Charmed baryon (q-q + Q) ⇔ X* (q + QQ) and W* (QQQ) systems• Heavy (heavier) quarks are key.

4

3q baryonMeson baryon

(Molecule)

Pentaquark

(Multi quark)

Q

q-q

q

Q Q

Q

QQ

X*

W*

Yc*

Experimental situations of X* and W* baryons

• Poor experimental data of X* and W*

• Systematics studies are essential.

• Narrow decay width expected

⇒ Chance to find and separate states• G ~ a few 10 MeV of X* < a few 100 MeV of L*/S*

• W* expects to have also narrow width.

⇒ Systematics measurements• Production and decay

＊K- beam is effective tool to produce multi-strangeness baryons.• High-momentum beam: 5-10 GeV/c

• Large yield: s = ~1 mb order and high-rate beam

⇒ Systematic studies by E50 spectrometer• Beam measurement is essential for missing mass and decay measurement

5

E50 spectrometer

1.0E+04

1.0E+05

1.0E+06

1.0E+07

1.0E+08

1.0E+09

0 2 4 6 8 10 12 14 16 18 20

[GeV/c]

Design Intensity [Hz] in spill (2.0 sec)

High-momentum beam line for 2ndary beam

• High-intensity beam: > 1.0×107 Hz p (< 20 GeV/c)• Unseparated beam: p/K/pbar : MHz K beam ⇔ 100 MHz p beam

• Beam timing measurement: Start timing

⇒ “Bottleneck” to increase beam intensity

6

@ 15 kW loss

0.E+00

5.E-03

1.E-02

2.E-02

2.E-02

0 2 4 6 8 10 12 14 16 18 20

[GeV/c]

K/p & pbar/p ratio

K-/p-

pbar/p-

2%p-

pbar

K-

MHz K beam

High-momentum beam line for 2ndary beam

• High-intensity beam: > 1.0×107 Hz p (< 20 GeV/c)• Unseparated beam: p/K/pbar : MHz K beam ⇔ 100 MHz p beam

• Beam timing measurement: Start timing

⇒ “Bottleneck” to increase beam intensity

7

1.0E+04

1.0E+05

1.0E+06

1.0E+07

1.0E+08

1.0E+09

0 2 4 6 8 10 12 14 16 18 20

[GeV/c]

Design Intensity [Hz] in spill (2.0 sec)

p-

pbar

@ 15 kW loss

MHz K beam

K-

• High-rate capability

• Good time resolution

High-rate Cherenkov timing detector: R&D status

Fine segment test

Signal processing method test

T0 detector

• Segment by Acrylic (PMMA)

⇒ Cross shape: X-type• Cherenkov angle direction

• Suppress time spread

• Both edge readout• ×2 light yield

• 3-mm width segment + MPPC• MPPC amplifier: ~10 ns width

⇒ Time resolution: DT ~ 50 ps(rms)• 3 MHz/segment ⇒Achieved• No position dependence＊Akaishi master thesis

＊Limit: ~3 MHz/3-mm segment

⇒ < 1-mm width fine segment• Handle 100 MHz beam

9

Front view

Beam

T0 detector

• Segment by Acrylic (PMMA)

⇒ Cross shape: X-type• Cherenkov angle direction

• Suppress time spread

• Both edge readout• ×2 light yield

• 3-mm width segment + MPPC• MPPC amplifier: ~10 ns width

⇒ Time resolution: DT ~ 50 ps(rms)• 3 MHz/segment ⇒Achieved• No position dependence＊Akaishi master thesis

＊Limit: ~3 MHz/3-mm segment

⇒ < 1-mm width fine segment• Handle 100 MHz beam

10

3 mm < 1 mm

Front view

30 MHz by 3-mm segment 100 MHz by 1-mm segment

Simulation: Radiator width dependence• Fine segment simulation

• Geant4: Optical photon• Realistic parameters: PMMA, MPPC and so on

• Normalization of # of p.e.• 3-mm radiator light yield data⇒ Reflection probability of PMMA: 99.5%

• Light yield is decreased by fine segments.• ~16 p.e. @ 0.5 mm

• Smaller loss of fast components• Small number of reflections

• Resolution is expected to be kept.

＊Production by company • Cut from one PMMA board

⇒Actual fine segment test

11

× < 50 ps

× < 100 ps

× < 150 ps

× < 200 ps

All P.E.

Single edge P.E.

Simulation image

Fine segments of Cherenkov radiator

• Radiators can be fixed by Silicone sheet with glue and some wires.

12

3.0 mm 1.0 mm 0.5 mm

Cross section

Test experiment @ LEPS

• Purposes

• Fine segment test

• 1.0 mm and 0.5 mm segments

• New amplifier test

• Signal processing test

• Schottky Barrier Diode filter circuit

• Integration circuit for TOT

• Time resolution evaluation by MIP• e± from g-ray conversion

• RF timing reference: DT ~14 ps

• Time walk correction by pulse height• Data taking: DRS4 and HUL HR-TDC

• MPPC (s13360-3050PE) conditions: Vov = +7V

• One p.e. pulse height: ~70 mV (amp gain: ×18.8)

13

Beam

Reference counters

New amplifier

(Test type)

T0 radiator

Number of photoelectrons

• Average: ~20 p.e.

• Light yield tendency of both edge is consistent.

14

Xup Xdown

Xup Xdown

Xup Xdown

3.0 mm

1.0 mm

0.5 mm

T. Akaishi, Master thesis

Beam

Number of photoelectrons (measured)

Xup

Xdown

20 mm high from center

Number of photoelectrons

• Average: ~20 p.e.

• Light yield tendency of both edge is consistent.

15

Xup Xdown

Xup Xdown

Xup Xdown

3.0 mm

1.0 mm

0.5 mm

T. Akaishi, Master thesis

Beam

Number of photoelectrons (measured)

Xup

Xdown

＊Handwriting curbs

Number of p.e.: @ +20 mm

• Sum of both edge and its distribution are same.

⇒ Collection of Cherenkov lights w/o surface loss

16

Xup Xdown

Xup Xdown

Xup Xdown

3.0 mm

1.0 mm

0.5 mm

38.7± 0.4 p.e.

(s = 7.8 p.e.)

Sum of both Edge

40.4± 0.5 p.e.

(s = 7.8 p.e.)

38.6± 0.7 p.e.

(s = 8.0 p.e.)

+

+

+

Time resolution: @ Vth = 3.5 p.e.

• All data: Similar time resolution of ~45 ps(rms).• Time resolution is kept. = Same light yield

＊3.0 mm ⇒ 0.5 mm: ×6 higher counting rate• 3 MHz/3 mm @ 30 MHz ⇒ 3 MHz/0.5 mm @ 180 MHz

17

3.0 mm 1.0 mm 0.5 mm

＊RF DT ~14 ps(rms) subtracted

30

38.7± 0.4 p.e.

Sum of both Edge

40.4± 0.5 p.e.

38.6± 0.7 p.e.

R&D issues of T0 detector

1. Ringing suppression

• Effects to time resolution by pile-up signal

• Time resolution: 43 ps⇒ 54 ps @ High-rate condition

⇒ Schottky Barrier Diode (SBD)

was used as kind of filtering methods.

2. TOT measurement

• Time-walk correction

by Time-Over-Threshold (TOT) method

• Width = (Leading edge – Trailing edge)

• Only TDC measurement without ADC

• Dead-time less digitalization for streaming DAQ

18Pileup event

Ringing

Trigger timing

Time-Over-Threshold (TOT) method

Schottky barrier diode: SBD• Kind of rectifier diode

• Quick responses• Smaller forward voltage: 100-200 mV level

＊Revered pulses and smaller pulses are suppressed.

⇒ Ringing suppression

• BAT63: Series connection to amplifier• Signals width: Same• Vout = 0.62×(Vin) – 70.0 (Minimum input: ~120 mV)

19I-V characteristic

MPPC amp

SBD circuit

Series connection

Time resolution: W/o and W/ SBD

• Similar time resolution of ~45 ps(rms).• W/ SBD data showed little worse resolution.• Smaller pulse height affected by noise

• SBD can be used as filter circuit. ⇒ High-rate test

20

＊RF DT ~14 ps(rms) subtracted

W/o SBD W/ SBD

＊Horizontal axis: ±0.3 ns

Ringing

Trigger timing

・ W/o SBD

×W/ SBD

Time-Over-Threshold method

• Straight forward method cannot correct time-walk of leading edge.

• Differential circuit for narrow signal width

⇒Width is saturated in the higher pulse height region.

＊Extract pulse height information from width

⇒ Integration by slow shaping

• SBD + slow shaping is essential.

• RC integration circuit

• t = 2.4 ns

• R = 51 W, C = 47 pF

• Signal width: ~15 ns

• Original: ~10 ns

21

VINVOUT

GND GND

R

C

Time resolution: TOT method

• Width analysis showed similar resolution. @ Vth = 4.5 p.e.• Response is not completely understood by using amplifier w/ SBD and RC circuit.

⇒ Investigation of optimum time constant

22

・ Pulse height

×Width

＊Vth = 3.5 p.e. data

⇒ Correlation is almost same as of

Vth = 4.5 p.e. data.

⇒ It is necessary to investigate.

By width

＊Tail component

＊RF DT ~14 ps(rms) subtracted

By PH

To do＊To finalize R&D and production of Cherenkov beam timing detector

• Tests in early 2020 are affected by COVID-19.• Segmented detector test by EMPHATIC @ Fermilab (Postponed by November)

⇒ Test at other facilities (LEPS, ELPH ?)

• High-rate test @ ELPH ⇒ It will be performed in July(?) or October (?).

+ Test by mixed signal: Emulate pile-up events

• Investigate and optimization of filtering circuit• SBD and RC circuit

• Design of actual detector• Segment width size adjustment for beam profile

• Simulation study for fixing radiators with good filling rate

• MPPC array (TSV type) for assembling fine segments

• Publication plan• X-shape Cherenkov detector: 1st draft is under preparation.

• High-rate measurement: Signal processing, TOT and so on

23

8 ch MPPC board

TSV MPPC array (1 mm)

Summary

• Investigation of effective degree of freedom for hadrons• Systematic study: Charmed baryon ⇔ X and W

• K- beam @ J-PARC High-p beam line• High-intensity K- beam ⇒ Dominant p in unseparated beam

• High-rate capability of beam timing detector: Fine segment beam detector

• R&D of Cherenkov timing detector• X-shape Acrylic radiator with thin width: 0.5 mm, 1.0 mm, 3.0 mm

• Test experiment @ LEPS

⇒ Time resolutions of ~45 ps(rms) were kept by using fine segment radiators.

• Availability of much higher counting rate beam: ×6 higher rate

• 3 MHz/3 mm @ 30 MHz ⇒ 3 MHz/0.5 mm @ 180 MHz

• Filtering method test for suppressing ringing effects by SBD

• High-rate test is necessary for finalizing R&D.

• TOT method test: SBD + Integration circuit

• It was found that measurement without ADC can be performed.

＊To finalize R&D and production of Cherenkov beam timing detector• K- beam intensity is as high intensity as possible at J-PARC high-momentum beam line.

⇒ To drive investigation of multi-strangeness baryons

24